Paper Citation Record · LEDGER

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization

As of 8 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 0 inbound Pith citation observations for arXiv:2607.18622.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2607.18622 v2

Coverage vector

measured 20 of 20 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-03T01:56:59.274728Z

measured 20 of 20 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

20 of 20 outbound references displayed

verified exact0
verified fuzzy0
unresolved18
parse uncertain0
malformed identifier2
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation d190f417-d402-481d-9572-87cb55831802 · outbound

This paper cites Query- based adversarial prompt generation.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Query- based adversarial prompt generation

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.447334Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 80d543ce-a7cf-4d4a-87ae-9f896c9f3eb3 · outbound

This paper cites Defending Against Indirect Prompt Injection Attacks With Spotlighting.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Defending Against Indirect Prompt Injection Attacks With Spotlighting

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.562647Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b86d549a-b4ad-470b-944e-d3a7316805d3 · outbound

This paper cites Attention tracker: Detecting prompt injection attacks in llms.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Attention tracker: Detecting prompt injection attacks in llms

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.643584Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3501e32d-a96c-4b6a-a0a4-4e0fefef94f1 · outbound

This paper cites Baseline Defenses for Adversarial Attacks Against Aligned Language Models.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Baseline Defenses for Adversarial Attacks Against Aligned Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.779480Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 19ea5d7d-369c-41fb-a598-312101913ad9 · outbound

This paper cites Pubmedqa: A dataset for biomedical research question answering.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Pubmedqa: A dataset for biomedical research question answering

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.853340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d38109ac-5807-47a6-87d7-53d253a4dca3 · outbound

This paper cites Instruction boundary: Quantifying biases in llm reasoning under various coverage.arXiv preprint arXiv:2509.20278,.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Instruction boundary: Quantifying biases in llm reasoning under various coverage.arXiv preprint arXiv:2509.20278,

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.972654Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5c7fa244-e9a5-4b67-a8fe-9d9116eea9da · outbound

This paper cites Tree of attacks: Jailbreaking black-box llms automatically.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Tree of attacks: Jailbreaking black-box llms automatically

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.065795Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation abc6bdd9-fa6d-4439-9573-ffd07e8ee868 · outbound

This paper cites Granite Guardian.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Granite Guardian

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.170447Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bcf92fe2-d32a-46de-a81d-f63e803f28b0 · outbound

This paper cites an unresolved cited work.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Unresolved cited work

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.277806Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 54c0e337-67a5-4452-8931-222af3d5dcea · outbound

This paper cites Cats Confuse Reasoning LLM: Query Agnostic Adversarial Triggers for Reasoning Models.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Cats Confuse Reasoning LLM: Query Agnostic Adversarial Triggers for Reasoning Models

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.553796Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9e597876-3897-42da-9cf1-594c97327fea · outbound

This paper cites LiveBench: A Challenging, Contamination-Limited LLM Benchmark.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization LiveBench: A Challenging, Contamination-Limited LLM Benchmark

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.763709Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c0eb748f-b430-4ab7-8cd1-cce3180d2e47 · outbound

This paper cites Black-box optimization of llm outputs by asking for directions.arXiv preprint arXiv:2510.16794,.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Black-box optimization of llm outputs by asking for directions.arXiv preprint arXiv:2510.16794,

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.893921Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 97eef707-615d-401b-a73e-58e2ffbae2cb · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.999868Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e7ca6615-69b6-4fa1-a270-9e4f1e036bea · outbound

This paper cites ,L(r) u o nu ! ,n u =max (1, round(λu)).

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization ,L(r) u o nu ! ,n u =max (1, round(λu))

Reference 19

Resolution: malformed identifier; no resolver link, observed 2026-08-03T01:56:59.163517Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 770b62f9-b048-4541-8acc-276f0a246960 · outbound

This paper cites CPInj achieves the highest Target ASR while maintaining perfect format compliance.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization CPInj achieves the highest Target ASR while maintaining perfect format compliance

Reference 20

Resolution: malformed identifier; no resolver link, observed 2026-08-03T01:56:59.274728Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation dd902880-98df-4544-b3eb-4b03dcac666c · outbound

This paper cites AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization AdvPrompter: Fast Adaptive Adversarial Prompting for LLMs

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:58.444116Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 266ef3f7-791e-4b9d-9e96-ed75a01509e0 · outbound

This paper cites Can textual gradient work in federated learning? InThe Thirteenth International Conference on Learning Representations, 2025a.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Can textual gradient work in federated learning? InThe Thirteenth International Conference on Learning Representations, 2025a

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.302491Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7a804f1-e460-4a19-8d8d-6b18ea3ec16c · outbound

This paper cites Folio: Natural language reasoning with first-order logic.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Folio: Natural language reasoning with first-order logic

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.378176Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d25f5654-317f-419d-969b-08a285c3e634 · outbound

This paper cites Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.704085Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3e37472e-aaf8-46c9-a9af-a051b615573b · outbound

This paper cites Jailbreaking Black Box Large Language Models in Twenty Queries.

CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization Jailbreaking Black Box Large Language Models in Twenty Queries

Reference 2026

Resolution: unresolved; no resolver link, observed 2026-08-03T01:56:57.266792Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.